Title:
DATA AMOUNT SUFFICIENCY DETERMINATION DEVICE, DATA AMOUNT SUFFICIENCY DETERMINATION METHOD, DATA AMOUNT SUFFICIENCY DETERMINATION PROGRAM, LEARNING MODEL GENERATION SYSTEM, TRAINED LEARNING MODEL GENERATION METHOD, AND TRAINED LEARNING MODEL GENERATION PROGRAM
Document Type and Number:
WIPO Patent Application WO/2021/260922
Kind Code:
A1
Abstract:
Provided is a data amount sufficiency determination device capable of determining the sufficiency of the data amount of training data with higher accuracy. A data amount sufficiency determination device according to the present disclosure comprises: a time-series data acquisition unit which acquires time-series data; a data division unit which divides the time-series data into a plurality of pieces of subsequence data; a data set generation unit which generates a plurality of subsequence data sets, each of which is a set of subsequence data; a feature amount calculation unit which calculates the feature amount of subsequence data; a probability distribution generation unit which generates a probability distribution of feature amounts for each subsequence data set; and a determination unit which determines whether or not the probability distribution has converged.
Inventors:
MASUZAKI TAKAHIKO (JP)
NASU OSAMU (JP)
NASU OSAMU (JP)
Application Number:
PCT/JP2020/025227
Publication Date:
December 30, 2021
Filing Date:
June 26, 2020
Export Citation:
Assignee:
MITSUBISHI ELECTRIC CORP (JP)
International Classes:
G06N99/00
Domestic Patent References:
| WO2015151558A1 | 2015-10-08 |
Foreign References:
| JP2016133895A | 2016-07-25 | |||
| JP2009135649A | 2009-06-18 |
Attorney, Agent or Firm:
MURAKAMI, Kanako et al. (JP)
Download PDF: